To calculate the cost of renewable energy under the RES, BHI used data from the Energy Information Administration, a division of the U.S. Department of Energy, to determine the percent increase in utility costs that Michigan residents and businesses would experience. This calculated percent change was then applied to calculated elasticities, as described in the STAMP® modeling section.
We collected historical data on the retail electricity sales by sector from 1990 to 2010 and projected its growth through 2025 using its historical compound annual growth rate (see Graphic 5).[*] To these totals, we applied the percentage of renewable sales prescribed by the Michigan RES. By 2015, and in all subsequent years, renewable energy sources must account for 10 percent of total electricity sales in Michigan.[36] (For the “Michigan Energy, Michigan Jobs” initiative simulation, renewable energy sources must account for 25 percent of total electricity sales by 2025).[37]
Next, we projected the growth in renewable sources that would have taken place absent the RES. We used the EIA’s projection of renewable energy sources by fuel for the Reliability First Corporation/Michigan through 2025 as a proxy to grow renewable sources for Michigan. We used the growth rate of these projections to estimate Michigan’s renewable generation through 2025 absent the RES. [38]
We subtracted our baseline projection of renewable sales from the RES-mandated quantity of sales for each year from 2011 to 2025 to obtain our estimate of the annual increase in renewable sales induced by the RES in megawatt-hours. The RES mandate exceeds our projected renewables in all years (2013 to 2025). This difference also represents the maximum number of MWh of electricity from conventional sources that are avoided, or not generated, through the RES mandate. We will revisit this shortly. Graphic 5 contains the results.
Graphic 5: Projected Electricity Sales, Renewable Sales and 10 Percent RES Requirement (Thousands of Megawatt-Hours)
To estimate the cost of producing the additional extra renewable energy under an RES against the baseline, we used estimates of the LEC, or financial breakeven cost per MWh, to produce the electricity.[39] However, as outlined in the “electricity generation cost” section above, the EIA numbers provide a rather optimistic picture of the cost and generating capacity of renewable electricity, particularly for wind power. A literature review provided alternative LEC estimates that were generally higher, and capacity factors that were lower, for renewable generation technologies than the EIA estimates.[†] We used these alternative figures to calculate our “high” LEC estimates and the EIA figures to calculate our “low” cost estimates and the average of the two to calculate our “medium” cost estimates. Graphic 6 displays the LEC and capacity factors for each generation technology.
Graphic 6: LEC and Capacity Factors for Electricity Generation Technologies
*These figures represent a weighted average of the estimates for both onshore and offshore wind. Onshore wind is weighted more heavily, since it is more likely to be used.
We used the 2016 LEC for the years 2010 through 2018 to calculate the cost of the new renewable electricity and avoided conventional electricity, assuming that from 2010 through 2016, the 2016 LEC would underestimate the actual costs, while from 2017 through 2018, the 2016 LEC would slightly overestimate the actual costs. We assumed that the differences would, on balance, offset each other. For 2019 and 2020 we used the 2020 LEC. The assumption is that LEC will decline over time due to technological improvements.
We used the EIA’s reference case scenario for all technologies. We adjusted the 2016 LECs to 2025 by using the percentage change in the capital costs from 2015 to 2025, since capital costs often represent the largest component of the cost structure for most technologies. For the technologies for which the EIA does not forecast LECs in 2020, we used the average of the 2016 and 2025 LEC calculations, assuming a linear change over the period.
Once we computed new LECs for the years 2020 and 2025, we applied these figures to the renewable energy estimates for the remainder of the period.
For conventional electricity, we assumed that the technologies are avoided based on their costs, with the highest-cost combustion turbine avoided first. For coal and gas, we assumed they are avoided based on their estimated proportion of total electric sales for each year. Although hydroelectric and nuclear are not the cheapest technologies, we assume no hydroelectric or nuclear sources are displaced since most were built decades ago and offer relatively cheap and clean electricity today.
To determine the impact of the RES standard in a given year, we calculated the amount of renewable energy the RES would require that year and compared it to our renewable energy baseline sales for that year; the difference represents the renewable sales attributable to the RES policy. We then determined which renewable energy source(s) would be used to meet the renewable energy sales attributable to the RES and calculated the additional renewable energy costs by using the LEC(s) for the relevant energy source(s).
The increased total costs in renewable energy lead to decreased total costs in conventional energy, since less conventional energy would be needed and sold. The decrease in conventional energy production is not as large as the increase in renewable energy production, however. Wind power and solar power in particular are intermittent (as reflected in their relatively low capacity factors), and it would still be necessary to keep backup conventional energy sources online and ready to meet any sudden electrical demands that renewable sources could not instantly provide. To estimate the share of conventional energy that would still be running as backup, we used a ratio of the renewable energy capacity factor to the conventional energy capacity factor.[‡]
Graphics 7, 8 and 9 below display the results of our medium-, low- and high-cost calculations for the 10 percent RES respectively. We converted the aggregate cost of the RES into a cost-per-kWh by dividing the cost by the estimated total number of kWh sold for that year. For example, for 2015 under the medium-cost scenario above, we divided $951 million by 111,853 million kWh for a cost of 0.85 cents per kWh.
Graphic 7: Medium-Cost Case of 10 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
Graphic 8: Low-Cost Case of 10 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
Graphic 9: High-Cost Case of a 10 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
Graphics 10, 11 and 12 below display the results of our medium-, low- and high- cost calculations for the 25 percent RES, respectively. We converted the aggregate cost of the RES into a cost-per-kWh by dividing the cost by the estimated total number of kWh sold for that year. For example, for 2025 under the medium-cost scenario above, we divided $2.551 billion by 132,405 million kWh for a cost of 1.93 cents per kWh.
Graphic 10: Medium-Cost Case of 25 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
Graphic 11: Low-Cost Case of 25 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
Graphic 12: High-Cost Case of a 25 Percent RES Mandate from 2013 to 2025 (Thousands of 2010 Dollars)
[*] “Electric Power Monthly: Table 8. Retail Sales, Revenue, and Average Retail Price by Sector, 1990 Through 2010,” (U.S. Energy Information Administration, 2012), (accessed Sept. 4, 2012). The historical compound growth rate was calculated independently for each sector — residential, commercial, industrial and transportation — using the years for which data were available. These independent rates were then used to project sales for each sector in subsequent years, with the projected total annual retail sales calculated as the sum of the projected annual sector sales. This calculation produces a projected annual compound growth rate of approximately 1.68 percent between 2013 and 2025.
[†] For coal, gas and nuclear generation we used the production cost estimates from the International Energy Agencies, Energy Technology Analysis Programs, “Technology Brief E01: Coal Fired Power, E02: Gas Fired Power, E03: Nuclear Power and E05: Biomass for Heat and Power,” (April 2010 http://www.iea-etsap.org/web/Supply.asp (accessed February 2012). To the production costs we added transmission costs from the EIA using the ratio of transmissions costs to total LEC costs. For wind power we used the IEA estimate for levelized capital costs and variable and fixed O & M costs. For transmission cost we used the estimated costs from several research studies that ranged from a low of $7.88 per kWh to a high of $146.77 per kWh, with an average of $60.32 per MWh. The sources are as follows: Andrew Mills, Ryan Wiser, and Kevin Porter, “The Cost of Transmission for Wind Energy: A Review of Transmission Planning Studies,” Ernest Orlando Lawrence Berkeley National Laboratory, http://eetd.lbl.gov/ea/emp/reports/lbnl-1471e.pdf (accessed Sept. 19, 2012); Competitive Renewable Energy Zones (CREZ) Transmission Optimization Study, The Electric Reliability Council of Texas, April 2, 2008 http://www.ercot.com/news/presentations/2006/ATTCH_ A_CREZ_Analysis_Report.pdf (accessed December 2010); Sally Maki and Ryan Pletka, Black & Veatch, California’s Transmission Future, August 25, 2010, http://www.renewableenergyworld.com/rea/ news/article/2010/08/californias-transmission-future (accessed December 2011).
[‡] For example, if the RES will require 100 MWh more wind than would otherwise be produced, then that 100 MWh of wind will produced at the LEC for wind. Ideally, then 100 MWh of natural gas-based energy would no longer be needed, and the forgone costs would be computed at the LEC for natural gas. Since wind would require a backup, however, we would estimate the amount of natural gas energy production needed on standby by employing a ratio of the capacity factors of the two energy sources (using, for example, the mid-range estimates from Graphic 6): 0.269/0.86 * 100 MWh of natural gas = 31.3 MWh of natural gas energy production.